Work

CARLA-GymDrive: Autonomous driving episode generation for the Carla simulator in a gym environment

Year: 2024

Type: software-paper

Abstract: CARLA-GymDrive is a powerful framework designed to facilitate reinforcement learning experiments in autonomous driving using the Carla simulator. By providing a gymnasium-like environment, it offers an intuitive and efficient platform for training driving agents using reinforcement learning techniques. It includes features such as scenario configuration to ensure that the training/test suite is adequate without requiring any code. Additionally, it boasts other features such as custom sensor configuration and compatibility with training libraries like Stable-Baselines3. This tool aims to increase researchers’ productivity by abstracting them from the complex code of the simulator, allowing them to focus on their research.

Source: SoftwareX

DOI: 10.1016/j.softx.2024.101955

Authors: Ângelo Miguel Rodrigues Morgado, Nuno Gonçalo Coelho Costa Pombo

Institutions: University of Beira Interior, Instituto de Telecomunicações

Language: English

Topic: Reinforcement Learning in Robotics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $970

Funder: Fundação para a Ciência e a Tecnologia